Review
Social anxiety and emotion knowledge: A meta-analysis

https://doi.org/10.1016/j.janxdis.2012.09.005Get rights and content

Abstract

It has been proposed that social anxiety is associated with poor emotion knowledge (EK), although studies have revealed mixed results. The aim of the present paper was to systematically investigate the association between EK and both non-clinical and clinical social anxiety by means of meta-analyses. Systematic, electronic database literature searches were performed, and meta-analyses were conducted on 43 included studies. Results showed that social anxiety was negatively associated with EK. The strongest association was found between clinical levels of social anxiety and the ability to understand one's own emotions (intrapersonal EK). Regarding interpersonal EK, a subgroup analysis showed that social anxiety was more strongly associated with a decreased ability to understand complex emotions than to recognize basic emotions. No differences were found between patients with social anxiety disorder (SAD) and patients with other anxiety disorders. Although a large between study heterogeneity and differing methodologies may prevent any firm conclusions from being reached, the results indicate that poor EK may play an important role in SAD, and that it could be beneficial to target EK in the treatment of SAD.

Highlights

► Social anxiety is associated with reduced emotion knowledge. ► Patients with social anxiety disorder show very poor intrapersonal emotion knowledge. ► No differences on emotion knowledge between anxiety disorders were detected.

Introduction

Social anxiety disorder (SAD) or social phobia is characterized by an intense fear in and avoidance of social situations leading to clinically significant distress and/or interference with the individual's normal routines (DSM-IV; American Psychiatric Association, 2000). If an individual fears most social situations, the condition may be specified as generalized SAD. Social anxiety may be considered a dimensional construct, with shyness at the lower end of the scale, and SAD (especially in its generalized form) and avoidant personality at the upper end, with rather arbitrary diagnostic thresholds (Ruscio, 2010, Stein et al., 2000). SAD is a pervasive and often chronic condition which has a considerable impact on academic, occupational, and social functioning, as well as on overall quality of life (Kessler, 2003, Wittchen et al., 2000).

A core feature of SAD is maladaptive emotion regulation, especially in the form of situational avoidance, emotion suppression (Kashdan and Breen, 2008, Kashdan and Steger, 2006, Werner et al., 2011), and post-event rumination (Clark & Wells, 1995). Emotion knowledge (EK) or emotion understanding may play a key role in adaptive emotion regulation. The present paper will adopt the following definition of EK, modified from Eisenberg, Hofer, and Vaughan (2007): [the ability to] identify one's own and others’ experienced and expressed emotions, understand which emotions are appropriate in different contexts, and recognize the causes and consequence of emotions (p. 296). Although closely related to emotion regulation, EK is different from emotion regulation in that it concerns what an individual knows about emotions, not how he or she handles them. That is, understanding which emotion regulation strategy would be suitable for a given situation would be considered EK, whereas actually employing a certain strategy would not. It has been suggested that increased EK better enables the individual to exercise adaptive control over emotions (Izard et al., 2011), and previous studies have confirmed that increased EK is indeed associated with emotion regulation competence (e.g. Barrett et al., 2001, Izard et al., 2008). Thus, entering an emotional situation with sufficient EK about what one and others feel, why those feelings arose, and which actions would be appropriate in the situation, appears to help the individual in choosing helpful emotion regulation strategies. However, the association between EK and emotion regulation is likely bidirectional, and reduced EK may also be the result of problematic emotion regulation strategies. One may argue, that a strategy like emotion suppression prevents an individual from attending to the information that emotions hold, and thereby hindering the experience of and learning about emotions. In fact, emotion regulation has been found to predict a person's EK over time (Schultz, Izard, Ackerman, & Youngstrom, 2001).

EK may reflect intrapersonal knowledge about one's own emotions and inner feelings, as well as interpersonal knowledge about other peoples’ emotions and feelings. The present study will investigate both types of EK, perceiving them as two distinguishable constructs. The two could, however, be related, since the ability to understand what happens emotionally in other people may rely on the ability to simulate part of the emotion intrapersonally (Niedenthal, Barsalou, Ric, & Krauth-Gruber, 2005). EK can be both explicit and implicit (Wranik, Barrett, & Salovey, 2007), that is, it encompasses strategic and conscious knowledge as well as automatic and implicit knowledge. It is a component of emotional intelligence (Salovey & Mayer, 1990), and is related to constructs such as alexithymia (Sifneos, 1973) and theory of mind (Baron-Cohen, Leslie, & Frith, 1985). EK differs from emotional intelligence in that the latter also includes how the knowledge is applied; alexithymia typically does not include the interpersonal aspect of EK (cf. Bagby, Parker, & Taylor, 1994); while theory of mind is a broader concept including knowledge about several kinds of psychological material. However, although somewhat different, the constructs seem to capture a meta-understanding of emotional experience. In this present paper, EK is thus viewed as an umbrella-term, including various constructs that concern the ability to detect and understand one's own and others’ emotions. However, several other concepts in the literature on emotion are referred to as either EK or emotion regulation, without always clearly distinguishing the two. As already stated, the concept of EK does not include what the individual does with his or her emotions (i.e. emotion regulation). Furthermore, EK as defined in the present paper does not include concepts such as attention to or acceptance of emotion. Eisenberg et al. (2007) included the ability to attend to emotions in their definition of EK. Although the degree to which an individual pays attention to his or her emotions likely influences an individual's EK, attention to emotion has been found to be distinct from the understanding of emotions (e.g. Palmieri et al., 2009, Salovey et al., 1995), and the two phenomena are not always associated. For instance, Gohm (2003) identified different clusters of individuals based on intensity of, attention to, and clarity of emotions; while some individuals were categorized as “cerebral” (high clarity of and average attention to emotions), others were categorized as “overwhelmed (average attention to but low clarity of emotions). Based on such findings, the present paper did not include the ability to attend to emotions in the definition of EK. Likewise, the operationalization of EK in the present study does not encompass acceptance of emotion, which is better viewed either as an attitude toward emotions (e.g. believing that it is OK to feel anxious) or as an emotion regulation strategy (e.g. accepting emotions when they arise).

Studies on EK have used questionnaires as well as performance-based measures. Using questionnaire data has well-known limitations, such as reliance on introspection. One may further argue that it seems somewhat paradoxical to ask an individual to self-report his or her EK by basically asking them to utilize EK. Additionally, most questionnaires of EK ask for stable cross-situational tendencies, and do not concern the dynamic aspects of the construct, such as the specific situational manifestations. Therefore, questionnaires may only deal with the first part of the present definition of EK, that is, the ability to identify one's own and others’ emotions. Thus, they do not concern the understanding of which emotions are appropriate in different contexts, as the latter part of the definition holds.

Performance-based tasks may be better suited both to objectively tap into an individual's EK, and to capture the more dynamic aspects of EK. One example of a performance-based test is the Mayer–Salovey–Caruso Emotional Intelligence Test (MSCEIT; Mayer, Salovey, & Caruso, 2003), which asks the respondent to solve emotion related problems based on vignettes. This measure, compared to questionnaires, has been found to be a better predictor of real-time social competence in men (Brackett, Rivers, Shiffman, Lerner, & Salovey, 2006). Other studies have also found that measures of EK by questionnaires and performance tests are not strongly associated, suggesting that they may be tapping into different mental processes (e.g. Brackett and Mayer, 2003, Brackett et al., 2006, Brannick et al., 2009).

It has been proposed that a lack of EK may be present in many disorders, and prominently so in SAD (Mennin et al., 2007, Mennin et al., 2009, Summerfeldt et al., 2011, Turk et al., 2005). Theoretically, SAD has most often been associated with poor interpersonal EK, since interpersonal problems are a core characteristic of SAD. Individuals with SAD have been proposed to interpret other people's emotional expressions in a distorted, negative and personalized way (Beck & Emery, 1985), or to overlook them due to their self-focused attention (Clark & Wells, 1995). There has been less focus on the relationship between SAD and intrapersonal EK, although it has been suggested that social anxiety may also lead to problematic appraisals and management of one's own emotional state (Clark and Wells, 1995, Rapee and Heimberg, 1997).

Emotions hold a range of information that influence the formation (and possible change) of beliefs (Boden & Berenbaum, 2010). When that information is not attended to (e.g. due to self-focused attention), it may lead to inaccurate beliefs and misinterpretations about one's own and other's emotions. Boden and Berenbaum (2012) further found that, even when controlling for attention to emotions, poor EK was associated with subclinical paranoia and suspiciousness, which characterize some individuals with SAD (Horowitz et al., 1988, Kashdan et al., 2009). Furthermore, several studies have detected poorer social competences in SAD compared to healthy controls (e.g. Beidel et al., 1985, Beidel et al., 2010, Stopa and Clark, 1993). As stated above, poor interpersonal EK has been associated with social competence, and poor social competences in SAD may partly rely on reduced EK. Additionally, EK is supposed to progress through interaction with others (Wranik et al., 2007), and due to the social inhibition characterizing individuals with SAD (Stein & Stein, 2008), it could be hypothesized that they are especially unwilling to share their emotions with others, and therefore are less likely to learn about and to fully understand their own and other's emotions.

Results from empirical studies have often supported an association between social anxiety and decreased intrapersonal EK (e.g. Evren and Evren, 2007, Fukunishi et al., 1997, Summerfeldt et al., 2011), whereas results for interpersonal EK are mixed, with some studies finding increased (e.g. Melfsen and Florin, 2002, Winton et al., 1995), and others decreased (e.g. McClure and Nowicki, 2001, Montagne et al., 2006) interpersonal EK among socially anxious individuals.

Emotional valence could play a role in EK in SAD, since socially anxious individuals may be especially sensitive to signs of evaluation by others, or may be overwhelmed by their own negative emotions. Due to the human face's particular significance in expression of interpersonal liking or disliking, detection of emotions from facial expressions may be especially pertinent for socially anxious individuals. Results may also vary with differences in the complexity of the emotion detection task. The identification of basic, facial emotional expressions (anger, disgust, fear, joy, sadness, and surprise), with their unique physiological characteristics (cf. Ekman, 1992), may require less cognitive processing than understanding more complex emotions, such as jealousy, or chains of emotions in social settings (Ekman, 1992, Power and Dalgleish, 1997). With simple emotion recognition tasks it may thus not be possible to discriminate between anxious and non-anxious individuals.

The main aim of the present study was to investigate the association between social anxiety and both intrapersonal and interpersonal EK by means of a systematic review and meta-analyses. The following variables were planned to be explored as moderators: participant type (SAD vs. non-clinical social anxiety), age (children/adolescents vs. adults), emotional valence (positive vs. negative), type of data (questionnaire vs. performance), emotion complexity (basic vs. complex emotions), and type of stimuli (facial vs. non-facial). Finally, for studies with clinical samples, participants with SAD were compared with participants with other anxiety disorder diagnoses to determine the specificity of the social anxiety–EK relationship.

Section snippets

Literature searches

Electronic PsycINFO, Pubmed, and Embase searches were conducted to locate studies on EK and social anxiety from the earliest dates available through December 2011. Two searches were performed at different time points (August and December 2011); one by the first author and another independently repeated by a research assistant. Disagreements between the two literature searches, which could not be explained by time difference (in 7 cases), were solved by consensus. The disagreements could likely

Search results and characteristics of studies

Information flow on study selection from the second and latest search is provided in Fig. 1. After duplicates were removed, 496 studies were identified and screened based on abstract, and 113 were ordered for full-text evaluation. More than 75% of the studies (N = 383) were excluded, which was most likely due to the broad literature search, including various search terms of which not all were as likely to produce relevant hits (e.g. “emotion experience”). Of the 113 studies that were ordered for

Discussion

The main finding of the meta-analyses was that social anxiety was negatively associated with EK, both in its intrapersonal and interpersonal form, which is in line with recent theory that suggests reduced EK to be a central feature in anxiety disorders in general (Mennin et al., 2007), and in SAD in particular (Mennin et al., 2009, Turk et al., 2005). The strongest association was found between social anxiety and poor intrapersonal EK, with a large effect size (r = −.63) in the clinical studies

Conclusion

The present study is the first study to systematically review EK in SAD, and is based on a relatively large number of studies. The main finding was that social anxiety was associated with reduced EK, especially in its intrapersonal form. Regarding interpersonal EK, social anxiety was more strongly associated with a decreased ability to understand complex emotions than to recognize basic emotions. There was no significant difference between SAD and other anxiety disorders. Although the large

Acknowledgements

The authors would like to thank Dr. Robert Zachariae, Dr. Michael Borenstein, Dr. Thomas Nielsen, Jacob Piet, and Patrick O’Toole for their help during the preparation of the manuscript.

References (126)

  • L. Dannahy et al.

    Post-event processing in social anxiety

    Behaviour Research and Therapy

    (2007)
  • B.M. DePaulo et al.

    Social anxiety and social judgment: the example of detecting deception

    Journal of Research in Personality

    (1994)
  • I. Fukunishi et al.

    Secondary alexithymia as a state reaction in panic disorder and social phobia

    Comprehensive Psychiatry

    (1997)
  • S. Hampel et al.

    The relations between social anxiety and social intelligence: a latent variable analysis

    Journal of Anxiety Disorders

    (2011)
  • M. Jacobs et al.

    Association between level of emotional intelligence and severity of anxiety in generalized social phobia

    Journal of Anxiety Disorders

    (2008)
  • T.B. Kashdan et al.

    Social anxiety and positive emotions: a prospective examination of a self-regulatory model with emotion suppression and expression tendencies as moderators

    Behavior Therapy

    (2008)
  • T.B. Kashdan et al.

    When social anxiety disorder co-exists with risk-prone, approach behavior: investigating a neglected meaningful subset of people in the National comorbidity survey-replication

    Behaviour Research and Therapy

    (2009)
  • J.S. March et al.

    The multidimensional anxiety scale for children (MASC): factor structure reliability and validity

    Journal of the American Academy of Child and Adolescent Psychiatry

    (1997)
  • R.P. Mattick et al.

    Development and validation of measures of social phobia scrutiny fear and social interaction anxiety

    Behaviour Research and Therapy

    (1998)
  • D.S. Mennin et al.

    Delineating components of emotion and its dysregulation in anxiety and mood psychopathology

    Behavior Therapy

    (2007)
  • D.S. Mennin et al.

    Emotion regulation deficits in generalized anxiety disorder, social anxiety disorder and their co-occurrence

    Journal of Anxiety Disorders

    (2009)
  • J. Mohlman et al.

    Jumping to interpretations: social anxiety disorder and the identification of emotional facial expressions

    Behaviour Research and Therapy

    (2007)
  • J.D.A. Parker et al.

    The relationship between emotional intelligence and alexithymia

    Personality and Individual Differences

    (2001)
  • P. Philippot et al.

    Social phobics do not misinterpret facial expression of emotion

    Behaviour Research and Therapy

    (2005)
  • S. Quadflieg et al.

    Recognition and evaluation of emotional prosody in individuals with generalized social phobia: a pilot study

    Behaviour Research and Therapy

    (2007)
  • R.M. Rapee et al.

    A cognitive-behavioral model of anxiety in social phobia

    Behaviour Research and Therapy

    (1997)
  • C. Rieffe et al.

    Psychometric properties of the emotion awareness questionnaire for children

    Personality and Individual Differences

    (2007)
  • C. Rieffe et al.

    Emotion awareness and internalising symptoms in children and adolescents: the emotion awareness questionnaire revised

    Personality and Individual Differences

    (2008)
  • R. Rosenthal

    The file drawer problem and tolerance for null results

    Psychological Bulletin

    (1979)
  • T.M. Achenbach

    Manual for the teacher's report form and 1991 profile

    (1991)
  • C.M. Ale et al.

    Facial affect recognition and social anxiety in preschool children

    Early Child Development Care

    (2009)
  • American Psychiatric Association

    Diagnostic and statistical manual of mental disorders

    (2000)
  • D. Archer et al.

    The interpersonal perception task (IPT)

    (1988)
  • R. Banerjee et al.

    Social-cognitive factors in childhood social anxiety: a preliminary investigation

    Social Development

    (2001)
  • R. Bar-On

    Bar-on emotional quotient inventory: short technical manual

    (2002)
  • L.F. Barret

    Feelings or words? Understanding the content in self-report ratings of experienced emotions

    Journal of Personality and Social Psychology

    (2004)
  • L. Barrett et al.

    Knowing what you’re feeling and knowing what to do about it: mapping the relation between emotion differentiation and emotion regulation

    Cognition and Emotion

    (2001)
  • A.T. Beck et al.

    Anxiety disorders and phobias: a cognitive perspective

    (1985)
  • D.C. Beidel et al.

    A new instrument to assess childhood social anxiety and phobia: the social phobia and anxiety inventory for children

    Psychological Assessment

    (1995)
  • M.T. Boden et al.

    The bidirectional relations between affect and belief

    Review of General Psychology

    (2010)
  • M. Borenstein et al.

    Introduction to meta-analysis

    (2009)
  • M.A. Brackett et al.

    Convergent, discriminant and incremental validity of competing measures of emotional intelligence

    Personality and Social Psychology Bulletin

    (2003)
  • M.A. Brackett et al.

    Relating emotional abilities to social functioning: a comparison of self-report and performance measures of emotional intelligence

    Journal of Personality and Social Psychology

    (2006)
  • M.T. Brannick et al.

    Comparison of trait and ability measures of emotional intelligence in medical students

    Medical Education

    (2009)
  • M.A. Bruch et al.

    Shyness and sociability reexamined: a multi-component analysis

    Journal of Personality and Social Psychology

    (1989)
  • K.S. Button et al.

    Interpreting the social scene: threat detection and cost attribution in social anxiety [abstract]

    European Neuropsychopharmacology

    (2010)
  • D.W. Campbell et al.

    Happy but not so approachable: the social judgments of individuals with generalized social phobia

    Depression and Anxiety

    (2009)
  • A. Carré et al.

    Difficulties identifying feelings, alexithymic features and brain responses in social phobia trait: an FMRI study [abstract]

    European Psychiatry

    (2011)
  • F.C. Celikel et al.

    Alexithymia and anxiety sensitivity in Turkish depressive anxiety and somatoform disorder outpatients

    International Journal of Psychiatry and Clinical Practice

    (2007)
  • D.M. Clark et al.

    A cognitive model of social phobia

  • Cited by (68)

    • Emotions in social anxiety disorder: A review

      2023, Journal of Anxiety Disorders
    • Social anxiety and behavioral assessments of social cognition: A systematic review

      2022, Journal of Affective Disorders
      Citation Excerpt :

      The main finding of this review was that most studies examining the association between SA and higher-level social cognition found a statistically significant effect, with the majority finding a negative association (higher levels of SA relating to less social cognitive ability). These findings are consistent with O’Toole et al. (2013) who found a negative association between SA and complex emotions (higher-level social cognition) and suggest that greater SA is associated with impairment in higher-level social cognition. Findings related to lower-level social cognition were mixed, although the majority of studies examining the domain of social perception found a statistically significant negative effect of SA.

    View all citing articles on Scopus
    View full text